There is a plethora of styles for citations and bibliographies in scientific journals and books, even within a single field. They built upon traditions, which are usually not arbitrary but were once chosen with well intentions for the benefit of either author, typesetter (now software) or reader.
Although it’s probably a good idea to not break expectations in this regard in general, has there actually been UX research on what is the most suitable reference style for readers? In other words, what are their needs and how to satisfy them best? Are they universal or are there severe differences between fields and media, calling for different solutions?
Some user desires
As a reader when reading a reference in text (i.e. a citation or cite) I want to easily recognize papers and books I already know. “Author” and “author-year” citations tend to do that better than “acronym” and “numeric” labels, but a numeric or symbolic footnote marker which links to full bibliographic details at the bottom of the current page may be just as good or even better for that, but that would probably only happen on first occurrence.
Unknown references, I would like to be able to look up as easily as possible in a library catalog etc. using the data found in the bibliography. To do that, traditionally, I needed to know who wrote it, how the piece is called exactly and where or wherein it was published, maybe also when. Today, however, unique identifiers like DOI, ISBN and URL can do that. Sometimes, many articles or chapters are cited in a bibliography that all come from the same collection or book – with editor often differing from author – whose information is repeated many times which could be solved by cross-references.
I also want to find a citation quickly and easily in a bibliography at the end. Obvious and short sort keys should help with that, but I may also skim through the bibliography without coming from the main prose and would like to know when there are several works by the same author, for instance. I think pictures, i.e. book covers, conference logos, software icons or author photos, could help as additional cues.
The year (or a more precise date) of when something was published may help to assert contemporary relevance and historic connections, otherwise it’s mostly helping to distinguish different works by the same author in a concise way.
In some fields of science conference proceedings, in others journal articles or even classic books are used as primary references. Sometimes, e.g. in the humanities or legal studies, one needs to distinguish sources of knowledge from objects of study which often also are written records.
I sometimes find it insightful to see in the bibliography on which other pages a reference has been cited – more so in books than articles, though.